2. Planning the public finances 2.1 The Chancellor’s fiscal rules In 1998, the Chancellor outlined two fiscal rules to constrain his future tax and spending decisions. His stated rationale was to provide a credible framework to ensure that the state of the public finances was sustainable and that it did not impose an unfair financial burden on future generations:1 • Τhe golden rule states that the government will only borrow to fund investment. This implies that tax revenues should equal or exceed current (or non-investment) spending. In other words, the ‘current budget’ should be in balance or surplus. Sensibly, the golden rule only has to be met on average over the economic cycle and not every year. This allows the ‘automatic stabilisers’ of the tax and benefit system to operate. When output in the economy is running below trend, tax revenues are subdued and spending on social security benefits rises. This temporarily increases government borrowing and enhances consumers’ spending power. The reverse happens in a boom: tax receipts increase as a share of national income and spending on benefits falls. If the rule had to be met every year, the government would be forced to respond to an increased current budget deficit, due to weak economic activity, by raising taxes or cutting its expenditure plans. This could exacerbate the downturn.2 • Τhe sustainable investment rule states that public sector debt should remain at a ‘stable and prudent’ level, interpreted by the Chancellor as no more than 40% of national income. The aim is to avoid a situation in which the government has to devote an ever-increasing share of national income to servicing its debt. Over the longer term, this would become both politically and economically unsustainable. If the Treasury’s forecasts to overachieve the golden rule are met, then its plans to increase public sector net investment to 2.2% of national income will be consistent with the sustainable investment rule.3 Both objectives are reasonable rules of thumb but neither is necessarily optimal.4 The golden rule has intuitive appeal but may not guarantee intergenerational fairness. There is no guarantee that the flow of benefits from an investment project will be synchronised with the debt repayments that 1 For more details, see HM Treasury, Analysing UK Fiscal Policy, London, 1999 (www.hmtreasury.gov.uk/media//89A63/90.pdf). 2 It is worth noting that, while the automatic stabilisers operate in the right direction, the strength with which they do so is a function of the precise structure of the tax and benefit system and may not be optimal from the perspective of macroeconomic management. 3 Also assuming that the Treasury macroeconomic forecasts are accurate. 4 A more detailed discussion of the government’s fiscal rules can be found in C. Emmerson, C. Frayne and S. Love, The Government’s Fiscal Rules, Briefing Note no. 16, Institute for Fiscal Studies, London, 2003 (www.ifs.org.uk/public/bn16.pdf). © Institute for Fiscal Studies, 2004 8 Planning the public finances finance it. It is also difficult to determine which sorts of spending yield benefits for future generations and which do not. The distinction drawn by the golden rule between capital and current spending suggests that all capital spending will benefit future generations and all current spending benefits only the present one. Given the use of national accounts definitions, this is unlikely to be the case. For example, some current expenditure on education might be thought to have greater long-term benefits than some capital spending projects. While it is clearly desirable that government indebtedness should not increase explosively, it is also hard to argue on theoretical grounds that the ceiling of 40% of national income laid down under the sustainable investment rule is necessarily more sensible than, say, 30% or 50%. International and historical comparisons both indicate that a 40% debt ceiling seems cautious, but this does not necessarily make it optimal in any rigorous, analytical sense. Also, focusing on public debt levels may not provide an accurate measure of the long-term sustainability of a country’s public finances. Governments have a number of de facto future financial obligations that are not counted as public sector debt, ranging from risks they have underwritten (either explicitly or implicitly) to the requirement to provide benefits and pensions to future generations.5 On the other hand, future governments also have the ability to increase revenues in the future by raising taxes. It is not obvious how either of these factors should be taken into account in deciding whether the government’s financial position at any given time is sustainable. Assessing whether the golden rule is met Retrospective assessment of the golden rule: dating the current cycle One can only pass definitive judgement on whether or not the golden rule has been met in retrospect, by examining whether the current budget has been in balance or surplus over a complete economic cycle. Assessing in the midst of a cycle whether the golden rule is on course to be met over the whole cycle is complicated by the need to identify where the economy stands in the cycle at any given time. To do so, it is necessary to estimate the ‘output gap’, a measure of the distance between actual national income and the potential level that is consistent with stable inflation (also described as the ‘trend’ level). Unfortunately, the level of ‘potential’ output cannot be observed directly, only estimated by looking at the past relationship between output and inflation and projecting how it will change in the future. In addition, it is necessary to estimate when the current cycle will end. This depends on the size of the output gap, the rate at which the economy is expected to grow in the future and the rate at which the potential level of output is expected to grow.6 5 For a more in-depth discussion of the sustainability of the public finances, see HM Treasury, Long-Term Public Finance Report: Fiscal Sustainability with an Ageing Population, London, December 2003 (www.hm-treasury.gov.uk/media//555E2/longterm_fiscal_1to6_436.pdf). 6 For information about the Treasury’s view on the sustainable level of economic growth, see HM Treasury, Trend Growth: Recent Developments and Prospects, London, April 2002 (www.hm-treasury.gov.uk/media//D6678/ACF521.pdf). 9 Green Budget, January 2004 The estimate of the output gap from 1990–91 to 2008–09 that the Treasury published in the December 2003 Pre-Budget Report7 (PBR) is shown in Figure 2.1. The current cycle is assumed to have begun in 1999–2000 following a mini-cycle between the first half of 1997 and mid-1999. Output then fell below trend again in 2001–02. Stronger growth is forecast to close the negative output gap by mid-2006, bringing the current cycle to an end. As an approximation, the Treasury thinks that the current economic cycle consists of the seven financial years running from April 1999 to March 2006. Figure 2.1. HM Treasury estimates of the output gap 3.0 Percentage of trend output 2.0 1.0 0.0 -1.0 -2.0 -3.0 -4.0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Year Note: Actual output less trend output as a percentage of trend output (non-oil basis). Source: Chart A3, page 179 of the December 2003 Pre-Budget Report. The 2003 Blue Book contained significant upward revisions to estimates of national income in recent years. In light of the revisions, the Treasury has increased its estimate of trend annual growth in potential output between the first half of 1997 and the third quarter of 2001 from 2.61% in the April 2003 Budget8 to 2.94% in the December 2003 Pre-Budget Report.9 Despite this increase, the Treasury’s projection for trend growth from the third quarter of 2001 until the end of 2006 remains 2¾% a year, although the Treasury uses a more cautious 2½% to project the outlook for the public finances. If the Treasury had increased its estimate of trend output growth from 2001 to 2006 in light of the revisions, this would have meant that the economy was presently even further below trend than the Treasury believes. So the output 7 HM Treasury, Pre-Budget Report, Cm. 6042, London, December 2003 (www.hmtreasury.gov.uk/pre_budget_report/prebud_pbr03/prebud_pbr03_index.cfm). 8 HM Treasury, Budget 2003: Building a Britain of Economic Strength and Social Justice, London, April 2003 (www.hm-treasury.gov.uk/budget/bud_bud03/bud_bud03_index.cfm). 9 See table A2, page 177 of the December 2003 Pre-Budget Report. 10 Planning the public finances gap would be even bigger than the Treasury’s current estimate of –1.4%. This is already a relatively large estimate by the standards of other forecasters.10 Retrospective assessment of the golden rule: Estimating the current budget The golden rule implies that the aggregate current budget balance over the seven years of the economic cycle should be in balance or in surplus. In previous speeches, the Chancellor quantified the government’s room for manoeuvre in meeting the golden rule by aggregating the past and forecast cash values of the current budget over the cycle. So in his 2003 Budget speech, for example, he noted that ‘We meet our golden rule over the cycle – not just achieving a balance but with an estimated surplus at £32 billions’.11 Figure 2.2. Current budget surplus as a percentage of national income Actual HM Treasury December 2003 Pre-Budget Report forecast Percentage of national income 2.5 £19.3bn £20.4bn 2.0 1.5 £8.8bn 1.0 0.5 0.0 -0.5 –£5bn -1.0 –£8bn –£11.8bn -1.5 -2.0 –£19.3bn 1999–2000 2000–01 2001–02 2002–03 2003–04 2004–05 2005–06 Financial year Note: Measures exclude the windfall tax and associated spending. Sources: December 2003 Pre-Budget Report; HM Treasury, Public Finances Databank, November 2003, London, 2003 (www.hmtreasury.gov.uk/media//95B19/PublicFinancesDatabank241103.XLS). As a result of the increased borrowing forecast for this and the next two financial years, the Treasury significantly reduced its estimate of the likely aggregate surplus in the December 2003 Pre-Budget Report: ‘By 2005–06, when the current cycle ends under the assumptions used in these projections, 10 For instance, the OECD Economic Outlook no. 74, published in December 2003 (www.oecd.org/document/61/0,2340,en_2649_33733_2483901_1_1_1_1,00.html), put the output gap for the UK at –1.1% of national income in 2003 and –0.9% in 2004. The International Monetary Fund puts the output gap in the third quarter of 2003 at just –¾% (www.imf.org/external/np/ms/2003/121803.htm). In addition, the Governor of the Bank of England, Mervyn King, in the November Inflation Report press conference, stated that he believed that with respect to the output gap ‘it’s hard to believe that it’s a long way from zero given that we’ve seen very stable, flat unemployment, stable nominal earnings growth, very stable inflation and growth not far from trend’. 11 Chancellor of the Exchequer’s Budget Statement, 9 April 2003 (www.hmtreasury.gov.uk/budget/bud_bud03/bud_bud03_speech.cfm). 11 Green Budget, January 2004 the accumulated total surplus over the economic cycle will be £4½ billion’.12 Figure 2.2 shows the expected current budget surplus over every year of the current economic cycle that is currently believed to run from 1999–2000 to 2005–06. The size of the surpluses in the first three years of the economic cycle (1999–2000, 2000–01 and 2001–02) outweighs the size of the expected deficits from 2002–03 to 2005–06 by £4½ billion. As Figure 2.3 illustrates, the predicted margin by which the golden rule will be overachieved in the current cycle has declined steadily over the past two-anda-half years. In Budget 2001, two years into the economic cycle, the Treasury was projecting a cumulative surplus of around £100 billion, compared with the cumulative surplus of just £4½ billion that the Treasury is now predicting. Figure 2.3. Predictions of the cumulative cash current budget surplus from various Pre-Budget Reports and Budgets 120 100 Budget 01 £ billion 80 Budget 02 Budget 03 60 PBR 03 40 20 0 Mar-99 Mar-00 Mar-01 Mar-02 Mar-03 Mar-04 Mar-05 Mar-06 Note: The cumulative current budget surplus is ‘set’ to zero in March 1999, when the cycle starts. Source: HM Treasury, various Budgets and Pre-Budget Reports. In his statement to Parliament in December 2003, the Chancellor used a slightly different method of calculation, focusing on ‘the average annual surplus on the current budget as a percentage of GDP since the cycle began’.13 This method is more favourable because the economic cycle begins with surpluses and ends with deficits, as £1 of surplus at the start is a larger share of national income than a £1 of deficit at the end due to both inflation and real growth in the economy. By contrast, the ‘cumulative current budget surplus’ method regards each £1 of nominal current budget surplus or deficit as the same, whatever year that £1 was accumulated in.14 12 Paragraph B37, page 214 of the December 2003 Pre-Budget Report. 13 Paragraph 2.62 of the December 2003 Pre-Budget Report. 14 If the rationale behind the golden rule was that future taxpayers should not be affected by the current spending decisions made over the current economic cycle, then the correct test would be adding the nominal surpluses and not taking the figures as a share of GDP. 12 Planning the public finances Over the first four years of the cycle, the current budget averaged surpluses of 1.0% of national income a year. The Treasury predicts deficits of 1.7% this year, 0.7% in 2004–05 and 0.4% in 2005–06, reducing the average over the cycle to date to 0.2% by the time the cycle ends. On this basis, the Chancellor said, ‘we not only meet our first fiscal rule, the Golden Rule, that we balance the budget, but even on cautious assumptions we have an average annual surplus over the whole cycle of around 0.2 per cent of GDP – meeting our first rule in this cycle by a margin of £14 billion’.15 The figure of £14 billion can be reached by multiplying the expected cumulative surplus over the cycle (expressed as a percentage of national income) by expected national income at the end of the cycle, in 2005–06. Figure 2.4 shows what the implied surplus would have been using this methodology alongside that using the previous methodology, based on previous Treasury forecasts for the current budget surplus and national income to March 2006. If the Treasury had forecast the cumulative surplus on the current budget as a percentage of national income for the current economic cycle in this way in April 2001, it would have obtained a cumulative forecast of around 10% (or 1.4% a year) of national income, or £119 billion, based on what it then expected national income to be in 2005–06. Figure 2.4. Predictions of the implied total current budget surplus in 2005–06 using the expected sum of cash amounts and expected averages, from various Pre-Budget Reports and Budgets Summing cash amounts 140 1.4% Summing shares of national income 120 1.1% £ billion 100 80 60 0.5% 40 0.2% 20 0 Budget 01 Budget 02 Budget 03 PBR 03 Notes: The percentage above each column shows what the average current budget surplus as a share of national income was expected to be by the end of 2005–06 at the time of each publication. Both the percentages and the absolute sums are calculated using contemporary predictions of national income in future years, not the most up-to-date forecasts from the December 2003 PBR. Source: HM Treasury, various Budgets and Pre-Budget Reports. 15 Chancellor of the Exchequer’s Pre-Budget Report Statement, 10 December 2003 (http://www.hmtreasury.gov.uk/pre_budget_report/prebud_pbr03/prebud_pbr03_speech.cfm). 13 Green Budget, January 2004 As Figure 2.4 shows, the underlying pattern is the same whichever method of calculation is used: having forecast two-and-a-half years ago that the golden rule would be overachieved by a huge margin, the Chancellor now finds his room for manoeuvre to be much diminished, partly as a result of deliberate policy decisions and partly as a result of forecast errors and revisions. To some extent, this is to be expected. Aiming to overachieve the golden rule by a large amount early in the cycle might be regarded as prudent, given the considerable uncertainties in forecasting the path of the public finances over a number of years. But there is no reason for the Treasury to seek to have overachieved the golden rule ex post by a significant amount, unless public debt were at or near the 40% of national income ceiling and the government wished to create room for more investment. It is therefore reasonable that the forecast overachievement of the golden rule should decline as the end of the cycle draws nearer and as the uncertainties surrounding the out-turn over the remainder of the cycle diminish. Margins of caution and forecasting error Deciding how much caution the Treasury should seek to build into its forecasts for the public finances at different points in the economic cycle is a matter of judgement. It depends on how confident one wants to be at any given point that existing tax and spending policies would be consistent with meeting the golden rule when the cycle ends. This, in turn, depends on what the pattern of forecast errors in the past suggests about the likely direction and magnitude of forecast errors in the future, and on the scope the Treasury sees for countervailing policy adjustments if such errors materialise. The Treasury’s average absolute error in forecasting public sector net borrowing – the current balance plus net investment – one, two, three and four years ahead for the period from 1977–78 to 2002–03 is shown in Table 2.1. This shows that even one year ahead, the average absolute error is £11.6 billion in today’s prices. Table 2.1. Average errors in forecasting public sector net borrowing (PSNB), as a percentage of national income and in £ billion Time period Average error Average error (% of national income) (£bn) One year ahead 1.0 11.6 Two years ahead 1.7 18.7 Three years ahead 2.2 25.0 Four years ahead 3.1 34.0 Notes: Figures in £ billion are calculated assuming HM Treasury GDP forecast for 2003–04 of £1,111 billion. Average error corresponds to the average absolute error over the period 1977– 78 to 2002–03 for one year ahead, 1981–82 to 2002–03 for two years ahead, 1982–83 to 2002–03 (excluding 1996–97 to 1999–2000) for three years ahead, and 1983–84 to 2002–03 (excluding 1984–85 to 1986–87 and 1997–98 to 2000–01) for four years ahead. Figures are based on HM Treasury data on average absolute error as a share of out-turn revenues and are calculated as a share of national income using the average value of tax receipts as a share of national income over the relevant period. Sources: Table 2.8 of HM Treasury, End of Year Fiscal Report, London, December 2003 (www.hmtreasury.gov.uk/pre_budget_report/prebud_pbr03/assoc_docs/prebud_pbr03_adend.cfm); authors’ calculations. 14 Planning the public finances When it comes to meeting the golden rule over the current cycle, the Treasury may be particularly concerned that its one-year-ahead forecast errors have been moving in a negative direction since the beginning of the current cycle. As shown in Figure 2.5, the forecast was about 2% of national income too pessimistic for 1999–2000, ½% too pessimistic in 2000–01, 1% too optimistic in 2001–02 and 1½% too optimistic in 2002–03. If the Pre-Budget Report forecast of a current deficit of 1.7% of national income in 2003–04 turns out to be accurate, then the one-year-ahead forecast in last year’s Budget will also turn out to have been almost 1% of national income too optimistic. Past experience suggests that if a forecast is overoptimistic in one year, it is more likely than not to be overoptimistic in the following year too.16 Figure 2.5. One-year-ahead, five-month-ahead and actual current budget balances as a percentage of national income Previous year’s Budget Pre-Budget Report Actual Percentage of national income 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 1999-2000 2000-01 2001-02 2002-03 2003-04 Financial year Source: HM Treasury, various Budgets and Pre-Budget Reports. Errors in forecasting the cumulative budget surplus over an economic cycle can arise from three sources: • errors in forecasting the path and composition of economic activity (and therefore the impact of the automatic stabilisers); • errors in forecasting tax revenues and spending for any given level and composition of national income; and • errors in estimating the level and growth rate of trend output. Such errors can affect the estimated timing of the economic cycle and the assessment of the underlying strength of the public finances. Errors in forecasting economic growth are relatively unimportant in explaining the Treasury’s errors in forecasting the budget balance over a horizon of at 16 See chart 2.4 of HM Treasury, End of Year Fiscal Report, December 2003 (www.hmtreasury.gov.uk/pre_budget_report/prebud_pbr03/assoc_docs/prebud_pbr03_adend.cfm). 15 Green Budget, January 2004 least up to four years.17 The present year’s current budget deficit, for example, seems likely to be significantly larger than the Treasury predicted at the time of the April 2003 Budget even though the Treasury’s forecasts for economic growth have been relatively accurate. In recent Budgets and Pre-Budget Reports, the Treasury has taken pains to demonstrate that it plans the outlook for the public finances in a deliberately cautious way. First, although the Treasury believes that the UK’s trend rate of output growth is 2¾%, the figure of 2½% is used in projections of the public finances. The intention is to avoid repeating the experience of the late 1980s, when a move into budget surplus was misinterpreted by the Treasury and many other forecasters as a structural improvement in the public finances rather than the by-product of an unsustainable boom in economic activity. In previous Budgets and Pre-Budget Reports, the Treasury also ‘stress-tested’ the public finance projections against a ‘cautious case’ in which the level of trend output was 1% lower than the central estimate. Such a reduction in the trend level of output would increase the share of past budget surpluses attributable to the cyclical strength of economic activity and reduce the share attributable to the underlying health of the public finances. Having predicted in every Pre-Budget Report and Budget since November 199818 that the golden rule would still be met on this basis, the Treasury conceded in the December 2003 PBR that ‘the average surplus on the current budget in the cautious case is no longer positive’, arguing in mitigation that this level of caution was no longer necessary with the end of the cycle drawing closer.19 It is indeed reasonable to argue that if the government is focused on meeting the golden rule over the current cycle, there is less need for caution as the end of the cycle draws nearer and uncertainty about the path of the public finances over the remainder of the cycle diminishes. But it could be argued that the Treasury has used up too much of its room for manoeuvre too quickly. With two-and-a-half years of the current seven-year cycle still to go, and having made overoptimistic forecasts three years running, the Treasury now expects the golden rule to be overachieved over the whole cycle by an amount smaller than its average error when forecasting an annual budget deficit two years ahead. In addition, the Treasury concedes that the golden rule will no longer be met if the negative output gap is one percentage point smaller than its current estimate of 1.4% of potential output – at a time when the International Monetary Fund estimates that the output gap is in fact only 0.75% of potential output. So on the Treasury’s own figures, on the basis of past errors and uncertainties, it is possible that further tax increases or reductions in public spending might be necessary to meet the golden rule over the current cycle. But is it sensible to judge whether policy is consistent with the golden rule by focusing on what is essentially an arbitrary definition of the ‘current’ cycle? If 17 See table B13 of HM Treasury, Pre-Budget Report, Cm. 4076, London, November 1998 (archive.treasury.gov.uk/pub/html/prebudgetNov98/index.html). 18 See, for example, chart 2.3, page 25 of HM Treasury, Pre-Budget Report, Cm. 4076, London, November 1998. 19 Paragraphs 2.71–2.74, pages 38–39 of the December 2003 Pre-Budget Report. 16 Planning the public finances the government made an average-sized overoptimistic forecast of the budget deficit one year before the cycle was due to end, it would hardly seem sensible to announce tax increases or spending cuts of more than £10 billion for one year simply to ensure that the golden rule over that period was still on course to be met. Cycles are a never-ending process and any point in time could be seen as the end of one cycle and the beginning of the next. It is not obvious that with public sector debt still below 40% of national income, the amount the government can borrow between now and March 2006 should be determined by the size of the surpluses it has run since April 1999. This backward-looking approach places undue emphasis on the exact dating of the economic cycle and on the legacy of previous fiscal policy decisions and external shocks, rather than focusing on the future sustainability of the present policy stance. Forward-looking assessment of the golden rule The prominence the Chancellor has given to meeting the golden rule over the current cycle means that the credibility of his long-term commitment to it may be severely diminished if it is missed on this basis. But as a guide for sensible policy-making, it is nonetheless better to ask whether existing tax rates and spending plans are consistent with meeting the rule in the future, over some appropriate medium-term time horizon, than over the current cycle. In the case of monetary policy, the Bank of England does not try to ensure that consumer price inflation averages 2% between two dates. Rather, the Bank sets interest rates at a level that will be consistent with hitting the inflation target approximately two years ahead. The emphasis is on the stability of future inflation. To assess whether policy is consistent with the golden rule looking forward, one must evaluate the underlying health of the public finances. The underlying position is judged by adjusting the current budget balance for the state of the economy, i.e. by asking what the current budget balance would be if national income were at its sustainable level, with an output gap of zero. If this ‘cyclically adjusted’ balance is in deficit and expected to remain so on existing policies, then those policies are unlikely to be consistent with meeting the golden rule looking forward. The cyclically adjusted current balance is shown alongside the actual current balances in Figure 2.6. This shows that: • following the fiscal tightening of fiscal policy in the 1981 Budget, the deficits on the current budget seen in the early 1980s were due to poor economic performance rather than to a structural weakness in the public finances; • the move into surplus in the late 1980s reflected the economic boom rather than a structural improvement in the fiscal position; • fiscal policy was loosened in the run-up to the 1992 election, resulting in even larger deficits than would have occurred anyway as a result of the recession of the early 1990s; • fiscal policy was then tightened significantly and consistently from 1993 to 1999; and • fiscal policy was then loosened again after 1999, when the current Labour government began to increase public spending more rapidly. 17 Green Budget, January 2004 Figure 2.6. Current budget surplus as a percentage of national income 8 Percentage of national income Actual 6 HM Treasury December 2003 Pre-Budget Report forecast 4 Cyclically adjusted surplus 2 0 -2 -4 -6 -8 78–79 81–82 84–85 87–88 90–91 93–94 96–97 99–00 02–03 05–06 08–09 Financial year Note: Measures exclude the windfall tax and associated spending. Sources: December 2003 Pre-Budget Report; HM Treasury, Public Finances Databank, November 2003, London, 2003 (www.hmtreasury.gov.uk/media//95B19/PublicFinancesDatabank241103.XLS). Figure 2.7. Predictions of the cyclically adjusted current budget balance, taken from various Pre-Budget Reports and Budgets Percentage of national income 2.5 Budget 02 2.0 PBR 02 Budget 03 1.5 PBR 03 1.0 0.5 0.0 -0.5 -1.0 1999- 2000- 2001- 2002- 2003- 2004- 2005- 2006- 2007- 20082000 01 02 03 04 05 06 07 08 09 Financial year Source: HM Treasury, various Budgets and Pre-Budget Reports. The Treasury now expects fiscal policy to tighten again over the next five years, moving from a cyclically adjusted current budget deficit this year of 0.8% of national income (the largest since Labour came to power) to a surplus of 0.6% in 2008–09. As Figure 2.7 illustrates, over the last 18 months, the Treasury has been forced regularly to revise down its forecast for the cyclically adjusted current budget balance in 2003–04, but it has always 18 Planning the public finances maintained that on existing policies, the underlying fiscal position would be back in healthy surplus by the end of its forecasting period. On the face of it, this suggests that current tax rates and spending plans are more than consistent with meeting the golden rule looking forward. The projections in the PBR do not necessarily show the expected path that the Treasury will choose to take. For that, we have to wait until the Budget. But it has given little indication that it believes a smaller tightening would be appropriate. Whether or not the current policy stance is consistent with the golden rule looking forward depends on whether one believes that the fiscal position will strengthen as strongly and as quickly as the Treasury believes on the basis of existing policies, above and beyond any improvement arising from stronger economic growth. As we discuss in Chapter 3, this prediction may be unduly reliant on ambitious forecasts for tax receipts. Even if one accepts the Treasury’s forecasts, it is important to be clear what the assumption of ‘existing’ spending and tax policies that underlies them implies. For spending, it implies that the steady rise in government current expenditure (including depreciation) from around 37% to 40% of national income seen since April 1999 comes to a halt this year and that the share is held roughly constant at just under 40% over the forthcoming spending review period and into 2008–09. The absence of a further significant increase in current spending on this measure underlines the pressure facing the government to deliver noticeable improvements in the quality of public services on existing levels of resources (which we discuss in Chapter 7). By contrast, public sector net investment is expected to continue rising from 1.6% to 2.2% of national income. This implies that total public spending would continue to rise as a share of national income, but much less quickly than over the past five years. Only a small fraction of the rise in current spending is due to the economic cycle. The forecast rise in current spending as a share of national income from April 1999 to March 2009 is 2.9% of national income while the increase in the cyclically adjusted figure is 2.7% of national income. Figure 2.8 shows cyclically adjusted current receipts and current expenditure as shares of national income. (The difference between the two series is the cyclically adjusted current budget surplus shown in Figure 2.6.) It is clear that the recent increases in current spending largely reflect policy decisions, not economic circumstances. For taxation, following a sharp drop in the share of national income taken in tax between 2000–01 and 2002–03, the government is now expecting receipts to rise by more than 2% of national income over the next five years, exceeding 40% for the first time since the late 1980s. Figure 2.8 shows that little of the projected rise in current receipts, like that of current expenditure as a share of national income, reflects the impact of the expected cyclical recovery in economic activity. Adjusted for the economic cycle, current receipts are projected to rise from 38.3% to 40.4% of national income over the next five years. This partly reflects economic factors that are not captured in conventional cyclical adjustment, such as rebounds in receipts from taxes that were depressed by the stock market slump and an expected ‘rebalancing’ of the composition of economic growth towards elements that are relatively 19 Green Budget, January 2004 highly taxed (such as consumer spending). Over the long term, such factors should not be expected to affect the share of national income taken in tax. The cyclically adjusted increase in current receipts also reflects the government’s efforts to collect more of the tax that it believes it is due – for example, by cracking down on VAT fraud and avoidance. The government believes that in 2001–02, it should have collected about £10 billion more in VAT revenue than it did, given existing rates and rules.20 Figure 2.8. Cyclically adjusted current receipts and current expenditure as a percentage of national income Percentage of national income 42.0 Cyclically adjusted current receipts 41.0 Cyclically adjusted current expenditure 40.0 39.0 38.0 37.0 36.0 1996-97 1998-99 2000-01 2002-03 2004-05 2006-07 2008-09 Year Notes: Current expenditure includes spending to offset depreciation. The cyclically adjusted figures for public sector current expenditure and current receipts were obtained using the ‘ready reckoner’ formula in table A5 of HM Treasury, End of Year Fiscal Report, December 2003 (www.hmtreasury.gov.uk/pre_budget_report/prebud_pbr03/assoc_docs/prebud_pbr03_adend.cfm). Source: HM Treasury website and December 2003 Pre-Budget Report. The most important source of increased future revenue is fiscal drag (which we discuss in more detail in Chapter 4). The Treasury assumes in forecasting the public finances that tax allowances and thresholds rise in line with retail prices. But earnings typically rise more quickly, which means that this assumption about allowances and thresholds implies a continuous rise in the share taken in income tax as more people find larger proportions of their income being taxed at higher rates. The impact of this is shown in the PBR: the Treasury forecasts that gross income tax receipts will increase from 11.1% of national income in 2005–06 to 11.6% of national income in 2008–09 despite the economy being at trend and no other discretionary tax changes in the pipeline. Left unchecked, fiscal drag would see the share of national income taken in income tax rise steadily for a considerable time. 20 Source: HM Customs and Excise, Measuring Indirect Tax Losses, London, November 2002 (www.hm-treasury.gov.uk/media//F7505/admeas02-297kb.pdf). 20 Planning the public finances In a trivial sense, ‘existing’ policies therefore clearly imply that the golden rule will be met in the future, thanks to fiscal drag. But, as the Long-Term Public Finance Report that accompanied the PBR acknowledges, this is unrealistic, and the Treasury therefore assumes that revenues and their composition remain broadly unchanged as shares of national income over the longer term.21 This implies ‘a comprehensive form of “real indexation”’,22 which presumably means tax allowances and thresholds rising more in line with earnings than with prices. The Treasury may indeed believe that, over the short to medium term, exploiting fiscal drag is the most sensible way to raise the extra revenues that it requires to meet the golden rule looking forward. But one should be clear that this would be a policy choice and not an economically neutral assumption. There are other ways that the revenue could be raised. Overall, the level of current expenditure as a share of national income is now almost the same as it was in 1996–97, when Labour had just come to power. Given the Chancellor’s commitment to meet his golden rule, and the fact that current spending is forecast to remain at this level, this implies that revenues as a share of national income need to be higher than they were in 1996–97. So far, tax revenues have not been sufficient to finance the significant increase in public spending and continue to meet the Chancellor’s golden rule looking forwards. Borrowing has therefore been higher and fiscal policy looser. This has helped lift some of the burden of sustaining economic growth in the UK from monetary policy. But continuing this higher level of borrowing is not compatible with the Chancellor’s golden rule, so the share of national income taken in tax is set to rise significantly over the next five years to finance the increase in spending we are seeing. The PBR projections make it clear that current receipts will have to rise relative to national income if the Chancellor wishes to meet the golden rule with the margin of comfort he has sought in the past. Should the economy not deliver the expected increase in tax receipts, then announcements of further tax increases, or reductions in currently anticipated levels of public spending, would be needed for this to remain the case. Assessing whether the sustainable investment rule is met As long as the golden rule is met, then the sustainable investment rule seems unlikely to impose a binding constraint on the government’s ability to spend without announcements of further tax increases in the medium term. Figure 2.9 shows public sector net debt as a share of national income from 1974–75 to 2008–09. This has declined steadily from 43.7% of national income in March 1997, just before Labour took office, to 30.2% of national income in March 2002. Net debt is set to rise over the next six years to 35.5% of national income in March 2009, according to the Treasury’s plans for receipts and spending. But assuming that the golden rule is met, the government would still be able to spend an additional one-off 4.5% of national income on public 21 Source: Paragraph 6.17, page 50 of HM Treasury, Long-Term Public Finance Report: Fiscal Sustainability with an Ageing Population, London, December 2003 (www.hmtreasury.gov.uk/media//555E2/longterm_fiscal_1to6_436.pdf). 22 Source: ibid., paragraph 6.18. 21 Green Budget, January 2004 sector net investment before breaching the sustainable investment rule. Given that public sector net investment is forecast to rise to just 2.2% of national income in 2008–09, such a large additional investment spend is unlikely. Figure 2.9. Public sector net debt as a percentage of national income Percentage of national income 80 70 60 50 40 30 20 10 0 74–75 77–78 80–81 83–84 86–87 89–90 92–93 95–96 98–99 01–02 04–05 07–08 Financial year Source: Pages 204 and 236 of the December 2003 Pre-Budget Report. Capital spending under the Private Finance Initiative In addition to public sector capital investment, some capital spending is undertaken by private firms on behalf of the public sector by means of the Public Finance Initiative (PFI). Under the PFI, the public sector pays private firms a rental price for use of a capital asset that the private sector delivers. In the absence of the PFI, the public sector would have had to undertake the accumulation of the capital itself, leading to higher debt if this were financed through borrowing as would be allowable under the golden rule. Figure 2.10 compares the Treasury’s projections of public sector net debt with the path that would have been seen if all PFI deals since 1990–91 had been conventionally financed through borrowing. By March 2006, public sector net debt would be 3.4 percentage points of national income higher, leaving the ratio of public sector net debt to national income just 2 percentage points below the ‘stable and prudent’ ceiling of 40% of national income. Given that public sector net debt is forecast to rise from 34.6% of national income in 2005–06 to 35.5% in 2008–09, and that the value of capital spending carried out under the PFI is likely to increase between now and 2008–09, it seems possible that by the end of the current planning period, adding PFI spending could push public sector net debt above the 40% of national income ceiling. But the Chancellor might reasonably argue that if he had intended to include PFI spending, he would have set the ceiling higher.23 23 For a more detailed discussion see chapter 2, pages 13–17 of R. Chote, C. Emmerson and H. Simpson, The IFS Green Budget: January 2003, Institute for Fiscal Studies, London, 2003 (www.ifs.org.uk/gb2003/ch2.pdf). 22 Planning the public finances Figure 2.10. Public sector net debt (PSND) as a percentage of national income Percentage of national income 80.0 PSND plus capital spending carried out under the PFI 70.0 Public sector net debt 60.0 50.0 40.0 30.0 20.0 10.0 0.0 1990-91 1993-94 1996-97 1999-2000 2002-03 2005-06 Financial year Notes: The figures for capital spending under the PFI prior to 2003–04 are adjusted from a calendar-year basis. Source: Figures up to March 2003 are taken from chart 2.3, page 7 of HM Treasury, Departmental Investment Strategies: A Summary, Cm. 5674, London, December 2002 (www.hm-treasury.gov.uk/media//343A6/dis_whitepaper02.pdf). Figures thereafter are taken from table C18, page 272 of the 2003 Budget. Private Finance Initiative payments are not the only set of liabilities that do not count as public sector net debt that might be faced by the government. For example, the government has underwritten some bonds issued by London Continental Railway relating to the Channel Tunnel Rail Link but has not scored this as borrowing, on the basis that the probability that the guarantee will be called is low.24 There is also the issue of whether borrowing carried out by Network Rail should be considered similar to conventional government borrowing. This depends on the likelihood that Network Rail would fail without increased support and, in this eventuality, that the government would offer financial support to ensure that the company survived.25 A measure of public sector indebtedness that did incorporate all these contingent liabilities might well exceed 40% of national income. While this would breach the ceiling for public sector net debt under the sustainable investment rule, the arbitrariness of the 40% limit, and the low level of debt relative to other countries and historical standards, mean that this is unlikely to affect any fundamental assessment of the sustainability of the UK government’s finances. 24 See page 34 of HM Treasury, Public Private Partnerships: The Government’s Approach, The Stationery Office, London, 2000 (www.hm-treasury.gov.uk/mediastore/otherfiles/80.pdf). 25 For more information on the classification of Network Rail, see www.statistics.gov.uk/about/Methodology_by_theme/rail_network/default.asp. 23 Green Budget, January 2004 2.2 The Maastricht Treaty fiscal rules Alongside the two fiscal rules described earlier in the chapter, the UK is also subject to further constraints on its debt and deficit as a signatory of the Maastricht Treaty. According to a protocol of the Maastricht Treaty, general government gross debt must be kept at below 60% of national income, while countries should have a medium-term objective of running a budget balance. The general government deficit should not exceed 3% of national income except under exceptional circumstances. Both these rules measure deficits and debt on a different basis from the UK’s two fiscal rules and some caution needs to be taken in comparing them. Nevertheless, it is clear that the requirement that the general government deficit should be at or below zero over the medium term and that it not exceed 3% is a stronger requirement than the golden rule, its public sector equivalent in the UK’s fiscal rules. In particular, it does not allow for ongoing borrowing for investment purposes as the golden rule does. According to the December 2003 PBR forecasts, the deficit on the Maastricht basis is expected to be 3.3% in 2003–04 and 2.6% in 2004–05.26 So the UK will almost certainly be in breach of the Maastricht limit this year and, given the large average errors associated with planning the public finances (see Section 2.1), could breach the limits again next year. Although the UK reports deficits and debt levels to the European Commission twice annually under the Excessive Deficits Procedure of the Maastricht Treaty, there is no provision for any action to be taken by the Commission if the limits are missed. As such, the Chancellor is able to continue to borrow on the terms allowable by his own rules, despite the possibility of being in breach of the Maastricht criteria he has also signed up to. Until the UK decides to join the Euro (if, indeed, it ever does), its breach of the Maastricht criteria is of little consequence either to the Chancellor or to other member states. A decision by the UK government to join the Euro might, in principle, increase the importance of the Maastricht rules. In practice, countries have been able to join the Euro despite being in breach of these rules. Table 2.2 shows government borrowing and debt for the 15 EU countries in 2002 and 1998. It shows that while all Euro countries were within the 3% limit in 199827 – the last year before the final stage of Euro entry – a number of countries that joined the Euro in 1999 were in breach of the debt rule. It was argued that as long as they were able to demonstrate falling debt, they should not be denied entry to the single currency. Table 2.2 also shows debt and deficit levels for 2002. The countries that are Euro members are now bound by the Stability and Growth Pact, which contains the same debt and deficit rules but includes the provision for monetary fines by the European Council if these are breached. In 2002, both France and Germany were in contravention of the deficit rule, and they are 26 Although the Maastricht criteria are discussed on calendar-year terms, the UK is assessed according to its usual financial years. 27 With the exception of Luxembourg, which qualified as being close to the limit. 24 Planning the public finances expected to have missed it in 2003 and to miss it in 2004. Both have successfully avoided facing the fines despite not being able to claim the mitigating circumstances of a ‘severe’ recession. Table 2.2. Public finances across the EU as a percentage of national income 2002 Public balance –0.2 0.1 2.1 4.2 –3.1 –3.5 –1.2 –0.2 –2.3 2.5 –1.6 –2.7 0.1 1.3 –1.5 –0.4 Debt 67.3 105.8 45.5 42.7 59.0 60.8 104.7 32.4 106.7 5.7 52.4 58.1 53.8 52.7 38.5 59.1 1998 Public balance –2.4 –0.7 1.1 1.5 –2.7 –2.2 –2.5 2.4 –2.8 3.1 –0.8 –2.6 –2.7 1.8 0.2 –0.6 Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom Unweighted EU average Weighted EU average –1.9 62.3 –1.6 Unweighted Eurozone –0.7 62.5 –1.0 Weighted Eurozone –2.2 69.0 –2.2 Source: Eurostat, EC Economic Data Pocket Book, no. 07-0/8/2003 (europa.eu.int/comm/eurostat/Public/datashop/printcatalogue/EN?catalogue=Eurostat&product=KS-CZ-03-000-3A-N-EN). Debt 63.7 119.6 56.2 48.6 59.5 60.9 105.8 54.9 116.3 6.3 66.8 55 64.6 68 47.6 66.3 68.7 68.5 73.5 In any event, the decision not to censure France and Germany for their consecutive breaches of the 3% deficit limit means that the Stability and Growth Pact has, in effect, been suspended. How the rules might be applied or developed in the future is not yet known, but it is possible that there would no longer be pressure on the UK to reduce its borrowing because of its European commitments. 2.3 A third fiscal rule: more active stabilisation By requiring achievement of the golden rule over an economic cycle rather than every year, the current fiscal framework gives free rein to the automatic stabilisers. The government can borrow to finance current expenditure during economic downturns provided that it runs offsetting current budget surpluses when national output is running above potential. This helps to limit swings in economic activity, although stabilisation is primarily the responsibility of the Bank of England. The Bank in effect uses monetary policy to keep the output gap as near to zero as possible once inflation is on course to hit its target. 25 Green Budget, January 2004 Joining the Euro would mean accepting the interest rate set by the European Central Bank (ECB) for the Eurozone as a whole. The UK accounts for a relatively small proportion of Eurozone economic activity, so there is no guarantee that the rate set by the ECB would stabilise economic activity in this country. The Treasury is therefore exploring the possibility of more active use of fiscal policy to stabilise the domestic economy if the UK joins the Euro. Even if the UK does not join the Euro, a more active stabilisation policy could be considered desirable. One possibility would be to make the automatic stabilisers more powerful – for example, by increasing social security benefits or making tax rates more progressive (so that a given increase in economic activity produced a larger increase in tax payments than is currently the case). But this could conflict with the government’s other objectives for redistribution and/or incentives. Alternatively, the government could use fiscal policy more actively in a discretionary way to support activity during downturns and to dampen it during booms. But given the failures of past attempts at fiscal stabilisation, the government is keen for people to believe that such a policy would be symmetric – in other words, that the government would be as keen to tighten policy in the good times as it would be to loosen it in bad times. Fiscal Stabilisation and EMU, published by the Treasury alongside the assessment of the five Euro tests,28 proposes a third fiscal rule which would mandate a temporary fiscal policy response whenever economic activity diverged by a given amount from its sustainable level in either direction. Possible policy tools might include changes in VAT or excise duty rates, or – if the swings in the economy are being driven by the housing market – in rates of stamp duty for property transactions. Whether or not such a rule would be helpful depends on a number of factors, including: • whether or not explicitly temporary tax changes would have a significant or predictable impact on economic activity; • whether or not people would anticipate the triggering of the rule and alter their behaviour in a way that would exacerbate rather than ameliorate swings in activity; • whether or not the rule would be credible – and therefore whether people would believe that the ‘temporary’ tax changes would in fact be reversed when activity returned towards trend; • whether or not the government is confident enough of its estimates of the output gap to believe that the rule would be triggered at appropriate times; and • whether or not the operation of such a rule would be consistent with the government’s commitments under the Excessive Deficits Procedure of the Stability and Growth Pact. If the procedure did not permit greater 28 HM Treasury, Fiscal Stabilisation and EMU, London, June 2003 (www.hmtreasury.gov.uk/documents/the_euro/assessment/studies/euro_assess03_studherefordshire.cfm ). 26 Planning the public finances borrowing when economic activity is depressed, then the activation of the new rule might well be inconsistent with the budget deficit limits laid down in the procedure. 2.4 Conclusions If the Treasury’s public finance projections are correct, then both fiscal rules will be met over the economic cycle that is predicted to run from April 1999 to March 2006. But its room for manoeuvre is greatly diminished: the cumulative surplus on the current budget that the Treasury expects is just £4½ billion or £14 billion depending on how it is calculated. Either way, this is less than the average absolute error that the Treasury has made in the past in forecasting public borrowing two years hence. An upward revision to borrowing before March 2006 of the magnitude seen in recent projections would necessitate substantial tax increases or spending cuts if the golden rule is to be met over the current economic cycle. A better interpretation of whether current fiscal policy is consistent with the golden rule is to look at projections for receipts and spending in future years. The Treasury’s public finance projections suggest that over the period 2003– 04 to 2008–09, current spending will remain just below 40% of national income while receipts will grow from 38.1% of national income in 2003–04 to 40.4% in 2008–09. If these forecasts prove correct, then the golden rule will continue to be met looking forwards. Should the economy not deliver the expected increase in tax receipts as a share of national income, then the Chancellor will have to announce further tax-raising measures or cuts in his existing tentative spending plans to expect to meet the golden rule with the degree of comfort he has sought in the past. Robert Chote, Carl Emmerson, Christine Frayne and Sarah Love 27